摘要

This paper presents an arc-based formulation of hydropower reservoir operation problems for application to ant colony optimization algorithms (ACOAs). The search space of the hydropower reservoir operation problem is defined as a partially connected graph composed of a set of arcs connecting the nodes of the graph. Each arc is defined such that it represents a local operation policy of the reservoir. The advantages offered by the proposed formulation are twofold. First, the formulation allows for a straightforward definition of heuristic information, a useful component of the ant-based algorithms, which is not present in other possible formulations of the problem. Second and more importantly, the proposed formulation paves the way for the efficient exploitation of the incremental solution-building mechanism inherent in ant algorithms to explicitly enforce problem constraints. A series of constrained formulations of the ACOA are, therefore, developed and used for the solution of single and multi-reservoir operation problems. The results indicate the ability of the proposed methods, and in particular the constrained versions, to optimally solve large-scale hydropower reservoir operation problems.

  • 出版日期2011-7

全文